PAUL v Royal Wolverhampton NHS Trust [2024] UKSC 1, [2024] 2 W.L.R. 417 is a leading case that never was. The scene had been set with care. Paul was the first major tort case on psychiatric injury to reach Supreme Court level for a quarter-century. The Court of Appeal had itself (unusually) granted permission to appeal, indicating disquiet with the law: [2022] EWCA Civ 12, [2023] Q.B. 149 (noted [2022] C.L.J. 452). Seven judges were listed to hear the appeal. But none of the parties invited the Supreme Court to reconsider the governing precedents, centrally Alcock v Chief Constable of South Yorkshire [1992] 1 A.C. 310. Prudently so, as the majority judgment, delivered by Lady Rose and Lord Leggatt, thought it improper “to contemplate such a radical departure from settled law” (at [116]). The court (with Lord Burrows dissenting) dismissed all three claimants’ appeals.

The outcome was conservative, although conservative with a fresh twist. First, the court sought to characterise liability for “secondary victim” psychiatric injury as anomalous. Accordingly, the strict limits on liability were consistent with principle, rather than contrary to it. Second, and following on, the court laid down a further restriction. The secondary victim’s psychiatric injury must occur by witnessing an accident (i.e. an injury to the primary victim by “violent external means”, at [24]). This precluded liability in the appeals heard together in Paul. Their common fact pattern involved medical defendants’ negligent failure to diagnose and treat an illness, the patients consequently dying from their underlying untreated conditions, with their deaths being witnessed by the claimants (close family members).

Since Lord Scarman’s influential speech in McLoughlin v O’Brian [1983] 1 A.C. 410, the principled approach to psychiatric injury has been thought to require a duty of care whenever such injury would be reasonably foreseeable. (Lord Burrows maintains this in dissent at [144]: foreseeability is “the only truly principled solution”). Courts have, however, rejected this principled solution for “policy” reasons, the decisive case being Alcock (supra) emphatically affirmed in Frost v Chief Constable of South Yorkshire [1999] 2 A.C. 455 (the two Hillsborough disaster cases). But in Paul the Supreme Court has questioned this orthodoxy, in the standard case of claimants traumatised by witnessing another’s death or injury (which includes McLoughlin, Alcock, Frost and Paul itself).

The court emphasised at [2]–[4] that a claim cannot be based on another person’s death at common law (Baker v Bolton (1808) 170 E.R. 1033); albeit Lord Campbell’s Act 1846 (now Fatal Accidents Act 1976) is a statutory exception to the rule, and “secondary” psychiatric injury claims “a further limited” exception at common law. From this starting point, the court concluded (at [140]) that what principally needs justification is not limits on such liability, but allowing “secondary” claims at all. (No positive justification was advanced; it was simply “too well established to be called in question” – Alcock, p. 410 (Lord Oliver)). There are no special rules restricting recovery;

It was necessary to draw a line somewhere, to keep this liability “within reasonable bounds” – or the exception would supplant the rule (at [141]).

The court (at [58]) therefore rejected Lord Hoffmann’s “lament” that “the search for principle was called off in Alcock” (Frost, p. 511). However, the Supreme Court contradicted its rehabilitation of Alcock as “principled” by accepting (also at [58]) that the Alcock limits were “influenced by practical and policy considerations rather than purely analogical development of the law”. (Quite what policy considerations is unclear when, at [48], the court was unimpressed by most of those identified by Lord Steyn in Frost, pp. 493–94, and summarised at [47]). We have passed through the looking-glass, and it is Lord Scarman’s broad approach to liability in McLoughlin v O’Brian that must now be seen as anomalous, and contrary to principle. The painful, consciously artificial line-drawing in Alcock (see e.g. Lord Oliver, p. 410, on the illogicality of the Alcock rules) was not as unprincipled as the Law Lords themselves had thought. Not only was Lord Hoffmann mistaken about Alcock in Frost, but also Lord Goff (who at p. 487 thought Alcock “arbitrary” and “artificial”) and Lord Steyn (Alcock produced “a patchwork quilt of distinctions which are difficult to justify”: p. 500).

Does the rule in Baker v Bolton bear the weight placed on it here? It seems questionable. First, not all psychiatric injury claims are “secondary”. There may be no primary victim at all. The claimant may (inaccurately) fear that their child or spouse has apparently been killed when they remain happily unscathed. Or the claimant may be shocked by destruction of their own property, such as their home or a scholar’s life’s-work (cf. Attia v British Gas [1988] Q.B. 304). Cases involving stress at work, or communication of upsetting information (e.g. incorrect medical diagnoses), cannot engage the “rule” either. But even in archetypal “witnessing” cases, a psychiatric injury claim is not wholly derivative or parasitic, unlike claims under the Fatal Accidents Act. It has hybrid elements (as Lord Burrows explains in the fullest analysis of the question at [212]–[223]). Centrally, for example, psychiatric injury to the secondary victim must be foreseeable. Lord Burrows briefly concludes at [224]–[225] that even if tort law generally denies liability arising from another’s injury, “the need for a separate duty of care to be owed to the secondary victim may be said to restore the general rule”. Also, as Lord Carloway explains in his concurring judgment, Scotland never recognised the (English) rule in Baker v Bolton. But that did not prevent Lord Carloway’s agreement with the majority (“Nothing turns on this [Scots] speciality”, at [253]). Lord Burrows records (at [212]–[213]) that submissions on this “conceptual” point had been requested by the court itself, but no party (including the defendants) “appeared to regard this issue as being of central importance”.

The defendants’ central argument, rather, was that there must be an “accident” external to the primary victim. As seen, the majority accepted this submission and dismissed the claims accordingly. As Lord Burrows noted in his dissent, this precludes virtually any claims involving deaths from medical errors. Although only one authority was formally overruled (North Glamorgan NHS Trust v Walters [2003] P.I.Q.R. P16) Paul was “a departure from the reasoning in almost all of the reported medical negligence cases in this area” (at [250]). Lord Burrows suggested the “accident” requirement “turned the clock back” (apparently, that is, to discredited medical theories postulating a “shock” to the nerves). The majority, however, defended “accident’s” legal significance. It postulates a clear, discrete event providing straightforward answers to the time and place at which the claimant must be present as a witness in order to claim (see [108]), in contrast with ongoing or deteriorating ill-health (see [112]–[113]). Lord Burrows, by contrast, thought that the death of the primary victim was the relevant event in the three claims in Paul, to which it was straightforward to apply the Alcock rules (putting aside “variant” cases of serious, non-fatal illness) (at [198]–[200]). (His Lordship queried at [211] why medical negligence was anyway not an “accident” as defined by the majority, being external to the secondary victim).

Medical liability was deliberately restricted. The court based a doctor’s (etc) duty on his or her assumption of responsibility exclusively to the patient (here “primary victim”), only exceptionally extending to the patient’s family (contrast ABC v St. George’s NHS Trust [2020] EWHC 455 (Q.B.), noted [2020] C.L.J. 214). Thus, apparently, precluding Lord Burrows’s “separate duty of care … owed to the secondary victim”. The court reasoned it would go far beyond the scope of the responsibility assumed if doctors had to protect patients’ friends and relations against the patient’s upsetting illness. Witnessing a relative’s death was not “an insult to health from which we expect doctors to take care to protect us but a vicissitude of life which is part of the human condition” (at [139]). Yet must watching one’s child or father die as a result of medical negligence be shrugged off as mere “vicissitude”? The court itself accepted at [143] that “[t]he thought that these tragic events could have been avoided if the hospital or doctor had exercised due care must, as in every case of wrongful death, add further to the agony and perhaps anger that [the claimants] feel”. It dismissed their claims despite the usual profession of sympathy.

The common law normally develops incrementally. Frost curtailed that process. The House of Lords there realised that Alcock’s arbitrary lines would inevitably be eroded if logic and principle governed liability for psychiatric injury. Hence (for Lord Steyn in Frost, p. 500) “thus far and no further”. Lord Burrows, dissenting in Paul, thought Frost a “counsel of despair” (at [204]). In 1998 the Law Commission had recommended radical reform (Law Com No 249). The Government had declined to implement this report, thinking it “preferable for the courts to have the flexibility to continue to develop the law rather than attempt to impose a statutory solution” (quoted at [146]). Thus, Lord Burrows reasoned, the Government had explicitly “passed the baton back to the courts” (at [237]). He did not think his proposed solution in Paul involved any change in the law, but if it did it was “well within the traditional judicial role … [of] incremental development” (at [201]–[202]). By contrast the majority reaffirmed Alcock, doubting the propriety of compensating secondary victims of psychiatric injury in any situation. Not only was the Supreme Court (in consequence) unwilling to consider modest relaxation of the limits on liability, but it announced a novel restriction in medical negligence cases (a distinction which Lord Burrows seemed to think “superficial and unprincipled” (at [241])). In their essentials, the Alcock rules survive. Little has changed for litigants after Paul, albeit scholars will continue to debate what would be the best “principled” solution.